Deepeval evaluate folder

[cathbrooks]

In this final version of the evaluate folder, there is test_evaluate_live.py, which evaluates a query from the command line, and test_evaluate_tsv.py, which has added capabilities of evaluating gold datasets.

Just cleaning up readme files, deleting some old data, and adding a method that will evaluate queries (with optional human answers in an additional column) that are stored in a .tsv files (sometimes the output contains a comma). Should work otherwise exactly as before.

[vercel[bot]]

@cathbrooks is attempting to deploy a commit to the Eye on Surveillance Team Team on Vercel.

A member of the Team first needs to authorize it.